Express method for determining the probability of developing congenital pneumonia in full-term newborns

Abstract

According to the World Health Organization, 15 % of children under the age of five died of pneumonia worldwide in 2017. Objective: to create a mathematical model for determining the probability of developing congenital pneumonia in full-term newborns using the express method. A comprehensive examination was conducted in 116 full-term newborn infants, who were admitted to pediatric ward for newborn children and maternity physiological department of the State Institution “Republican Scientific Practical Center “Mother and Child” during the period from 2017 to 2019. A comparative analysis of 174 life history data, outcomes of previous pregnancies, complications of current pregnancy in mothers, clinical and routine laboratory methods used in newborns on day 1–2 of life was performed. Subsequently, the most significant factors associated with the development of congenital pneumonia in full-term newborns were determined. Based on binary logistic regression and receiver operating characteristic (ROC) analysis, a mathematical analysis of 387.4 million variable relationships was performed. The survey determined the most significant factors associated with the development of congenital pneumonia in full-term newborns: from the past obstetric-gynecologic and somatic history of the mother — a spontaneous abortion; complications of pregnancy — chronic intrauterine fetal hypoxia and chronic fetoplacental insufficiency; the results of clinical manifestations in the first days of life — the presence of respiratory failure. Based on the obtained data, a model was developed with a sensitivity of 90.6 %, specificity of 92.1 % and the area under the ROC curve = 0.930 ± 0.054, p < 0.001 (95% confidence interval 0.88–0.98). Threshold values were calculated using a predictive model: for spontaneous abortion — 0.836; for chronic intrauterine fetal hypoxia — 0.537; for chronic fetoplacental insufficiency — 0.533; the presence of respiratory failure — 0.939, which at threshold values of ≥ 0.53 of the ROC curve makes it possible to identify a group at high risk of developing congenital pneumonia among full-term newborns. The current mathematical model for determining the probability of developing congenital pneumonia in full-term newborns using the express method can be found on the website http://pneu.bsmu.by or https://www.bsmu.by (at the bottom of the page, there is Doctor/Pharmacist: Diagnosis of Congenital Pneumonia menu).

Keywords

• model
• risk factors
• full-term newborns
• congenital pneumonia